Anomaly Detection of Big Data Based on Improved Fast Density Peak Clustering Algorithm

Aiming at the problems of traditional clustering algorithms in the process of large amount of big data anomaly monitoring under the background of big data, a big data anomaly detection method based on improved fast density peak clustering algorithm is proposed. Set up a big data anomaly clustering framework, automatically select parameters and clustering centers, evaluate big data outliers through standardized local density and distance, and can obtain outliers, extract and calculate thresholds according to features, complete anomaly scheduling, and achieve anomaly detection. The clustering algorithm designed in this paper is used for example analysis, which shows that the algorithm designed in this paper can meet the needs of actual users and improve the detection accuracy of power big data outliers.